Abstract

Cybersecurity in educational settings faces ever-evolving challenges, demanding adaptive and context-aware solutions. This study presents a groundbreaking real-time implementation of a combined approach utilizing Deep Deterministic Policy Gradient (DDPG), OpenAI Gym, and Deep Q Networks (DQN) for detecting and preventing Denial of Service (DoS) and Distributed Denial of Service (DDoS) attacks on student performance. The AI-driven solution showcased remarkable efficacy, leading to a significant reduction in attacks from 3.5*10^7 to 10^4 within a few weeks. The system's rapid response, facilitated by DDPG and DQN, demonstrated its adaptability to emerging threats. Context-aware defense mechanisms within OpenAI Gym contributed to the precision in identifying and preventing attacks, ensuring a stable learning environment. The integration of a Smart Study Advisor (SSA) further enhanced the system's positive impact on student performance. This real-time success provides a tangible and measurable example of the potential of adaptive cybersecurity solutions in educational settings. The results underscore the practicality and effectiveness of the combined DDPG, OpenAI Gym, and DQN approach, offering promising
implications for future advancements in educational cybersecurity.